Re: [cmis] Proposal for "Mash-up/Browser" bindings

[Al Brown <albertcbrown@us.ibm.com>]

This was an interesting topic at the meeting in Basel. This came up from a couple of items:
1. For clients that work with either binding, information to store is different - URI vs ObjectId. Having a simple URI that returns an atom entry on specification of an objectid would be very useful
2. Including a JSON representation would also be very useful
3. Supporting/documenting how children + descendant resources could accept/handle form posts would be helpful
4. If we include #1 in the app service document, we could then extend to other more rpc'ish style URI mechanisms such as get folder by path, updateProperties, move, etc.

David N and David C were excited about it and they are going to go off and create a proposal for review by the TC in a month. I do not think it has to be a new binding. The Rest/Atom binding can be extended to address it.

I'd like to see at least #1 addressed since there are clients that are supporting both bindings.

-Al

Al Brown
Emerging Standards and Industry Frameworks
CMIS: https://w3.tap.ibm.com/w3ki07/display/ECMCMIS/Home
Industry Frameworks: https://w3.tap.ibm.com/w3ki07/display/ECMIF/Home

Office 714 327 3453
Mobile 714 263 6441
Email albertcbrown@us.ibm.com
CONFIDENTIAL NOTICE: The contents of this message, including any attachments, are confidential and are intended solely for the use of the person or entity to whom the message was addressed. If you are not the intended recipient of this message, please be advised that any dissemination, distribution, or use of the contents of this message is strictly prohibited. If you received this message in error, please notify the sender. Please also permanently delete all copies of the original message and any attached documentation.

David Nuescheler ---05/06/2009 02:53:43 PM---hi guys,

From:	David Nuescheler <david@day.com>
To:	Derek W Carr/Watson/IBM@IBMUS
Cc:	Randall_Craig@emc.com, cmis@lists.oasis-open.org, ethang@exchange.microsoft.com
Date:	05/06/2009 02:53 PM
Subject:	Re: [cmis] Proposal for "Mash-up/Browser" bindings

---

hi guys,

thanks a lot for the the thoughtful comments. i would definitely
be thrilled to have as many reviewers and contributors as possible.

i definitely appreciate the timing issues around this and i am happy to have
this as an optional deliverable of the specification and see how far we
get aligning things with the current schedule.

having said that, i would definitely like to state, that if we are actually even
thinking of introducing a server-sided proxy (as mentioned by derek) that
translates cmis into "another protocol" that can be consumed by a
browser to cover the most simplistic usecases we definitely missed
the mash-up usecase for cmis as a protocol by quite a bit.

regards,
david


On Wed, May 6, 2009 at 9:11 PM, Derek W Carr <dwcarr@us.ibm.com> wrote:
> I had brought a similar issue up at the previous interoperability events
> where we demonstrated our widgets functioning in IBM Mashup Center.
>
> There are a number of issues in the browser today that are challenging to
> work-around in order to do proper XML parsing of atom response documents
> [specifically around namespaces].  In my experience, it has not been
> painful sending atom documents to the server from the browser, but the
> complexity in receiving it can be problematic given the current state of
> browsers.  If we introduce a multi-part POST it would allow us to remove a
> server-side proxy for our browser clients to support document upload/edit
> scenarios.  There are some issues that we as a TC would need to resolve if
> we introduce a multi-part POST endpoint to support document upload/edit
> specifically around introducing mechanisms to prevent CSRF attack vectors.
>
> Given the set of issues I would agree with Ethan that it would be nice to
> see this targeted as a follow-on separate release.  I would like to assist
> in any future definition of this binding.
>
> Thanks,
> ---------------------------------------
> Derek Carr
> (919) 254-8592 (t/l 444)
> ---------------------------------------
>
>
>
>  From:       Randall_Craig@emc.com
>
>  To:         <ethang@exchange.microsoft.com>, <david@day.com>, <cmis@lists.oasis-open.org>
>
>  Date:       05/06/2009 12:58 PM
>
>  Subject:    RE: [cmis] Proposal for "Mash-up/Browser" bindings
>
>
>
>
>
>
> Quick question: is this about adding additional representation support
> to the same set of resources (i.e. Atom, JSON, XML, etc.), or is this
> really about adding another binding?
>
> -----Original Message-----
> From: Ethan Gur-esh [mailto:ethang@exchange.microsoft.com]
> Sent: Wednesday, May 06, 2009 9:52 AM
> To: David Nuescheler; CMIS List
> Subject: RE: [cmis] Proposal for "Mash-up/Browser" bindings
>
> David,
>
> This is a really interesting conversation topic, and one that I suggest
> we add to the agenda for next week's TC conf call.
>
> Personally, I think it would be really interesting to think about a
> JavaScript/JSON binding of a subset of the CMIS protocol that would
> facilitate easier creation of "mashup" applications as you describe.
>
> That said, my $0.02 here:
>
> We intentionally structured the CMIS specification that way we did so
> that we could easily add new protocol bindings independent of the domain
> model, and at this point I have a strong personal desire to see the CMIS
> domain model and the two current bindings get "released" sooner rather
> than later.
>
> I believe that getting this 3rd binding right would take a non-trivial
> amount of time, and that if we're going to do it we should do it in the
> same thoughtful way we've designed these bindings.
>
> So I love the concept, and I'd love to see the TC spin up a separate
> deliverable for this mashup/web binding that we can target for a
> separate (later) release.
>
> Thoughts?
>
> - Ethan.
>
> -----Original Message-----
> From: David Nuescheler [mailto:david@day.com]
> Sent: Wednesday, May 06, 2009 2:17 AM
> To: CMIS List
> Subject: [cmis] Proposal for "Mash-up/Browser" bindings
>
> Dear TC members,
>
> as part of the upcoming Apache Chemistry [1] project I started producing
> a JavaScript library that would allow to access the atom bindings and
> would allow a mash-up client (browser) to interact with a CMIS compliant
> repository. As you can see on the CMIS matrix [2], the JavaScript client
> was able to connect to a variety of repositories.
>
> I noticed that the atom bindings do not lend themselves to be consumed
> by a lightweight browser client and for example cannot even satisfy the
> very simple usecase of uploading a file from the browser into a CMIS
> repository. Even simple read operations require hundreds of lines of
> JavaScript code.
>
> Based on my impressions I started to have conversations with a number of
> other TC members during the plugfest and over lunch I had a discussion
> with Al and others at my table around the topic of adding an additional
> binding that was based on JSON GETs for reading and multi-part POSTs for
> writing. The goal of this binding would be to deliver on the mash-up
> usecase and allow for very simple interaction with current web browsers.
> from a functional standpoint these bindings don't necessarily need to
> cover the full spectrum of cmis functionality, but instead should allow
> the most simple and convenient access to the repository just to break
> down the entry barrier for web developers.
>
> I think we could call it the "mashup-" or "web-" or "browser-binding".
>
> It seemed that everybody I talked to was very excited, so I would like
> to volunteer to write and orchestrate a proposal (Dave Caruana
> volunteered to help me ;) ). I understand that we would be operating
> under a great time pressure based on our schedule but i am confident
> that having such a binding would make CMIS a true web 2.0 specification
> and hence would add a lot of value.
>
> Please let me know what you think.
>
> regards,
> david
>
> [1] http://incubator.apache.org/chemistry
> [2] http://liip.to/cmismatrix
>
> --
>
> web:  http://www.day.com/ http://dev.day.com
> twitter: @daysoftware
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
>
>
>
>



-- 
David Nuescheler
Chief Technology Officer
mailto: david.nuescheler@day.com

web:  http://www.day.com/ http://dev.day.com
twitter: @daysoftware

---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php